CN113610285A - Power prediction method for distributed wind power - Google Patents
Power prediction method for distributed wind power Download PDFInfo
- Publication number
- CN113610285A CN113610285A CN202110846795.8A CN202110846795A CN113610285A CN 113610285 A CN113610285 A CN 113610285A CN 202110846795 A CN202110846795 A CN 202110846795A CN 113610285 A CN113610285 A CN 113610285A
- Authority
- CN
- China
- Prior art keywords
- power
- wind
- prediction
- data
- wind power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims abstract description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 42
- 238000012549 training Methods 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 239000000178 monomer Substances 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000007619 statistical method Methods 0.000 abstract description 6
- 238000005259 measurement Methods 0.000 description 21
- 238000005516 engineering process Methods 0.000 description 11
- 238000013461 design Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Water Supply & Treatment (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a power prediction method of distributed wind power, which comprises the following specific steps: and collecting data required by power prediction of the wind power plants in the designated area, and grouping the wind power plants in the area through correlation among output power of each wind power plant. The power prediction method of the distributed wind power is characterized in that real-time meteorological data of a fan head anemoscope, a anemoscope and the like are accessed by taking a meteorological principle and an engineering principle as theoretical basis and combining a numerical calculation and statistical method, so that real-time meteorological data of each layer height in the whole field range and any position in the field are obtained.
Description
Technical Field
The invention relates to the technical field of new energy, in particular to a power prediction method of distributed wind power.
Background
The distributed wind power generation is located near an electric load center, aims at large-scale long-distance power transmission are not taken as the purpose, the generated power is connected to a power grid nearby, and the wind power is consumed locally, and the power prediction result precision of the distributed wind power station is improved by the technology for improving the power prediction result precision, so that the short-term and ultra-short-term power prediction precision of the wind power station with the distributed fan arrangement can be improved, and the wind power station has gained wide attention of domestic and foreign scientific research units in recent years. A research team develops a power prediction system of a distributed wind power plant through implementation of a technical subject of improving the precision of a power prediction result of the distributed wind power plant.
The research is carried out around the key scientific and technical problems of the power prediction system of the distributed wind power plant. The power prediction precision is researched by means of researching the power prediction precision by using a single fan for prediction, researching the use of various meteorological sources and integrated meteorology, comparing and optimizing with actually-measured meteorology, researching the combination of a time sequence and a neural network model and a space-time model, researching a simulation wind measurement technology of multi-layer high multi-meteorological-element of a distributed wind power plant and the like based on single-machine data; and finally obtaining a key technology for improving the accuracy of the power prediction result of the distributed wind power plant. Through research of the subject, reference is provided for application of a distributed wind power plant power prediction technology, and the method has important significance for promoting the wind power plant to improve prediction accuracy.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a power prediction method of distributed wind power, which is characterized in that power prediction precision is researched by means of researching and using a single fan for prediction, researching and using various meteorological sources and integrated meteorology, comparing and optimizing with actual-measured meteorology, researching combination of a time sequence and a neural network model and a space-time model, and researching a simulation wind measurement technology of multilayer high and multiple meteorological elements of a distributed wind power plant based on single-machine data; and finally, the key technology for improving the accuracy of the power prediction result of the distributed wind power plant is obtained, and the like, and the problems are solved.
(II) technical scheme
In order to achieve the purpose of obtaining real-time meteorological data of each layer height in the whole field range and at any position in the field by accessing real-time meteorological data of a fan head anemoscope, a anemoscope and the like and combining numerical calculation and a statistical method by taking a meteorological principle and an engineering principle as theoretical basis, and finally obtaining a key technology for improving the precision of a power prediction result of a distributed wind power plant by considering the design of a regional prediction model, matching representative wind resource measurement data and selecting different regions based on the distribution of fans to respectively perform power prediction modeling because a single prediction model cannot meet the precision requirement of the wind power plant with a large wind range, the invention provides the following technical scheme: a power prediction method of distributed wind power comprises the following specific steps:
1) collecting data required by power prediction of wind power plants in a designated area, and grouping the wind power plants in the area through correlation among output power of each wind power plant;
2) selecting each group of representative wind power plants according to historical observation meteorological data and topographic features, and selecting a virtual anemometer tower position in the representative wind power plants;
3) establishing a power prediction model representing the wind power plant according to the data obtained in the step 1, and calculating short-term prediction power and ultra-short-term prediction power of the power prediction model;
4) and (3) integrating the data of each single machine and the data required by prediction collected in the step (1), performing combined calculation by adopting an AdaBoost algorithm, and predicting the power prediction result of the wind power plant.
Preferably, the required data in step 1 includes historical wind speed data and historical power of all the individual wind turbines, and average wind speed data and actual power of the whole field.
Preferably, after the model is built in the step 3, the average wind speed data and the actual power data of the whole field are provided for at least 2 weeks on site after the system is put into test operation.
Preferably, the time resolution of the average wind speed data and the actual power of the whole field included in the data required in step 1 is 1min or 5min, and when the time resolution is 10min, the data needs to be processed for the second time.
Preferably, in step 2, the wind turbines are scattered, and the wind power plants are considered to represent, so that the predicted weather points are more than or equal to three.
Preferably, the maximum error of the daily prediction curve in the step 3 is less than or equal to 25%:
i is the number of points, n is 96 points, Pni is the ith point short-term power predicted value, and Pri is the ith point actual power;
the hourly harmonic mean accuracy is more than or equal to 75%:
i is the point number, n is 96 points, Pni is the ith point ultra-short-term power predicted value, Pri is the actual power of the ith point, real-time meteorological data of each layer height in the whole field range and any position in the field are obtained by taking the meteorological and engineering principles as theoretical basis, accessing real-time meteorological data of a fan head anemoscope, a anemoscope and the like and combining numerical calculation and statistical methods.
Preferably, in the AdaBoost algorithm in step 4, there are two weights, the first is the weight of each sample in the training set, which is called sample weight and is represented by vector D; the other is that each weak learning algorithm has a weight, represented by the vector α.
Let us assume a training set of n samples { (X)1,y1),(X2,y2),...,(Xn,yn) Initially, setting the weight of each sample to be equal, namely 1/n, learning the samples by using a first weak learning algorithm h1, and counting the error rate epsilon after learning is completed:
where # error represents the number of misclassified samples and # all represents the number of all samples, the error rate ε can be used to compute the weight of weak learning algorithm h 1:
after the first learning is completed, the weights of the samples need to be adjusted again, so that the samples which are wrongly weighted in the first classification can be repeatedly learned in the following learning:
wherein h ist(xi)＝yiIndicates correct training for the ith sample, ht(xi)≠yiIndicating a training error for the ith sample. Zt is a normalization factor:
Zt＝sum(D)
the second learning is carried out, and after t rounds of learning are carried out, t weak learning algorithms { h } are obtained1,h2,....,htAnd its weight { alpha }1,α2,....,αtAnd calculating t weak classifiers (h) for the new classification data respectively1(X),h2(X),....,ht(X), the final output result of the AdaBoost algorithm is as follows:
sign (x) is a symbolic function, an AdaBoost algorithm is a machine learning algorithm based on a Boosting idea, wherein AdaBoost is an abbreviation of Adaptive Boosting, AdaBoost is an iterative algorithm, and the core idea is to train different learning algorithms, namely weak learning algorithms, aiming at the same training set, and then to combine the weak learning algorithms to construct a stronger final learning algorithm.
Preferably, four monomer methods, namely a BP neural network, a GRNN neural network, an ELM extreme learning machine and an ORPC optimization regression power curve, are combined in the AdaBoost algorithm, the accuracy of virtual wind measurement data is verified according to the correlation between the wind turbine data and the virtual wind measurement data, the actual measurement data of a real-time wind measurement system plays a very critical role in evaluating the virtual wind measurement data, important basis is provided for virtual wind measurement revision and parameter adjustment, the optimization of the virtual wind measurement data is realized, the accuracy is improved, the error of a power prediction system is reduced, so that the power prediction accuracy is integrally improved, and because a wind power plant with a large wind range cannot meet the accuracy requirement by considering the design of a subarea prediction model, the representative wind resource measurement data is matched, the subarea prediction model is based on the distribution of the wind turbines, different areas are selected, and respectively carrying out power prediction modeling to finally obtain a key technology for improving the precision of the power prediction result of the distributed wind power plant.
(III) advantageous effects
Compared with the prior art, the invention provides a power prediction method of distributed wind power, which has the following beneficial effects:
1. the power prediction method of the distributed wind power is based on the theory of meteorology and engineering, real-time meteorological data of a fan head anemoscope, a anemoscope and the like are accessed, and a numerical calculation and statistical method is combined, so that real-time meteorological data of each layer height in the whole field range and any position in the field are obtained.
2. According to the power prediction method of the distributed wind power, because a single prediction model cannot meet the precision requirement in a wind power plant with a large wind range, a regional prediction model is designed, representative wind resource measurement data are matched, different regions are selected based on the distribution of fans by considering the regional prediction model, power prediction modeling is carried out respectively, and finally the key technology for improving the precision of the power prediction result of the distributed wind power plant is obtained.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
A power prediction method of distributed wind power comprises the following specific steps:
1) collecting data required by power prediction of wind power plants in a designated area, and grouping the wind power plants in the area through correlation among output power of each wind power plant;
2) selecting each group of representative wind power plants according to historical observation meteorological data and topographic features, and selecting a virtual anemometer tower position in the representative wind power plants;
3) establishing a power prediction model representing the wind power plant according to the data obtained in the step 1, and calculating short-term prediction power and ultra-short-term prediction power of the power prediction model;
4) and (3) integrating the data of each single machine and the data required by prediction collected in the step (1), performing combined calculation by adopting an AdaBoost algorithm, predicting the power prediction result of the wind power plant, wherein the combined prediction has higher prediction performance and less risk of extreme prediction error compared with single prediction.
The required data in the step 1 comprise historical wind speed data and historical power of all single fans, and average wind speed data and actual power of the whole field; after the model is built in the step 3 and the system is tested, providing the average wind speed data and the actual power data of the whole field for at least 2 weeks on site; the time resolution of the average wind speed data and the actual power of the whole field included in the required data in the step 1 is 1min or 5min, when the time resolution is 10min, the data needs to be processed for the second time, and because a power limiting time period exists, manual power limiting records need to be provided on site; in the step 2, as the fans are scattered and represent the wind power plant, the prediction weather points are more than or equal to three; the maximum error of the daily prediction curve in the step 3 is less than or equal to 25%:
i is the number of points, n is 96 points, Pni is the ith point short-term power predicted value, and Pri is the ith point actual power;
the hourly harmonic mean accuracy is more than or equal to 75%:
i is the number of points, n is 96 points, Pni is the ith point ultra-short-term power predicted value, Pri is the actual power of the ith point, real-time meteorological data of each layer at any position in the whole field range and the field can be obtained by taking the meteorological and engineering principles as theoretical basis, accessing real-time meteorological data of a fan head anemoscope, a anemoscope and the like and combining numerical calculation and statistical methods, and because the wind power field with larger wind range cannot meet the precision requirement by a single prediction model, different regions are selected by considering the design of a sub-region prediction model and matching wind resource measurement data with representativeness, and the sub-region prediction model is based on the distribution of fans, power prediction modeling is respectively carried out, and finally the key technology of improving the precision of the power prediction result of the distributed wind power field is obtained; in the AdaBoost algorithm in the step 4, two weights are provided, wherein the first weight is the weight of each sample in a training set, is called as a sample weight and is represented by a vector D; the other is that each weak learning algorithm has a weight, represented by the vector α.
Let us assume a training set of n samples { (X)1,y1),(X2,y2),...,(Xn,yn) Initially, set the weight of each sample to beEqual, namely 1/n, is learned by using a first weak learning algorithm h1, and after learning is completed, statistics of the error rate epsilon is carried out:
where # error represents the number of misclassified samples and # all represents the number of all samples, the error rate ε can be used to compute the weight of weak learning algorithm h 1:
after the first learning is completed, the weights of the samples need to be adjusted again, so that the samples which are wrongly weighted in the first classification can be repeatedly learned in the following learning:
wherein h ist(xi)＝yiIndicates correct training for the ith sample, ht(xi)≠yiIndicating a training error for the ith sample. Zt is a normalization factor:
Zt＝sum(D)
the second learning is carried out, and after t rounds of learning are carried out, t weak learning algorithms { h } are obtained1,h2,....,htAnd its weight { alpha }1,α2,....,αtAnd calculating t weak classifiers (h) for the new classification data respectively1(X),h2(X),....,ht(X), the final output result of the AdaBoost algorithm is as follows:
wherein sign (x) is a symbolic function, AdaBoost algorithm is a machine learning algorithm based on Boosting idea, wherein AdaBoost is an abbreviation of Adaptive Boosting, AdaBoost is an iterative algorithm, the core idea is to train different learning algorithms, namely weak learning algorithms, aiming at the same training set, then the weak learning algorithms are combined to construct a stronger final learning algorithm, in order to construct a strong learning algorithm, firstly the weak learning algorithm needs to be selected, and the weak learning algorithm is continuously trained by using the same training set to improve the performance of the weak learning algorithm, the AdaBoost algorithm combines four monomer methods, namely BP neural network, GRNN neural network, ELM extreme learning machine and ORPC to optimize a regression power curve, the accuracy of the virtual wind measurement data is verified according to the correlation between the wind turbine data and the virtual wind measurement data, the real-time wind measuring system has a very key effect in evaluating the virtual wind measuring data, provides an important basis for virtual wind measuring revision and parameter adjustment, realizes the optimization of the virtual wind measuring data, improves the accuracy, reduces the error of the power prediction system, and integrally improves the power prediction precision.
The invention has the beneficial effects that: the method for predicting the power of the distributed wind power system obtains real-time meteorological data of each layer height in the whole field range and any position in the field by accessing real-time meteorological data of a fan head anemoscope, a anemoscope and the like and combining a numerical calculation and statistical method by taking a meteorological theory as a theoretical basis, and finally obtains a key technology for improving the precision of the power prediction result of the distributed wind power system by considering the design of a sub-region prediction model, matching representative wind resource measurement data, selecting different regions based on the distribution of fans and respectively carrying out power prediction modeling on the sub-region prediction model and considering the fact that a single prediction model cannot meet the precision requirement of the wind power system with a larger wind range, and verifying the accuracy of the virtual wind measurement data according to the correlation between the fan data and the virtual wind measurement data, wherein the actual measurement data of the real-time wind measurement system plays a very important role in evaluating the virtual wind measurement data, important basis is provided for virtual anemometry revision and parameter adjustment, optimization of virtual anemometry data is achieved, accuracy is improved, errors of a power prediction system are reduced, and therefore power prediction precision is integrally improved; combined prediction has higher prediction performance and less risk of extreme prediction error than monomer prediction.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. A power prediction method of distributed wind power is characterized by comprising the following specific steps:
1) collecting data required by power prediction of wind power plants in a designated area, and grouping the wind power plants in the area through correlation among output power of each wind power plant;
2) selecting each group of representative wind power plants according to historical observation meteorological data and topographic features, and selecting a virtual anemometer tower position in the representative wind power plants;
3) establishing a power prediction model representing the wind power plant according to the data obtained in the step 1, and calculating short-term prediction power and ultra-short-term prediction power of the power prediction model;
4) and (3) integrating the data of each single machine and the data required by prediction collected in the step (1), performing combined calculation by adopting an AdaBoost algorithm, and predicting the power prediction result of the wind power plant.
2. The method for predicting the power of the distributed wind power as claimed in claim 1, wherein the required data in step 1 includes historical wind speed data, historical power of all the individual wind turbines, and average wind speed data and actual power of the whole field.
3. The method as claimed in claim 1, wherein after the model is built in step 3, the average wind speed data and the actual power data of the whole wind farm are provided for at least 2 weeks on site after the system is put into operation.
4. The method for predicting power of distributed wind power as claimed in claim 3, wherein the time resolution of the actual power and the average wind speed data of the whole field included in the data required in step 1 is 1min or 5min, when the time resolution is 10min, the data needs to be processed for the second time, and since there is a power-limiting period, a manual power-limiting record needs to be provided on the site.
5. The method as claimed in claim 1, wherein in step 2, the wind turbines are dispersed and represent a wind farm, so that the predicted weather point is equal to or greater than three points.
6. The method according to claim 1, wherein the maximum error of the daily prediction curve in step 3 is less than or equal to 25%:
i is the number of points, n is 96 points, Pni is the ith point short-term power predicted value, and Pri is the ith point actual power;
the hourly harmonic mean accuracy is more than or equal to 75%:
i is the number of points, n is 96 points, Pni is the ith ultra-short-term power prediction value, and Pri is the ith actual power.
7. The method according to claim 1, wherein in the AdaBoost algorithm in step 4, there are two weights, the first is the weight of each sample in the training set, called sample weight, and is represented by vector D; the other is that each weak learning algorithm has a weight, represented by the vector α.
Let us assume a training set of n samples { (X)1,y1),(X2,y2),...,(Xn,yn) Initially, setting the weight of each sample to be equal, namely 1/n, learning the samples by using a first weak learning algorithm h1, and counting the error rate epsilon after learning is completed:
where # error represents the number of misclassified samples and # all represents the number of all samples, the error rate ε can be used to compute the weight of weak learning algorithm h 1:
after the first learning is completed, the weights of the samples need to be adjusted again, so that the samples which are wrongly weighted in the first classification can be repeatedly learned in the following learning:
wherein h ist(xi)＝yiIndicates correct training for the ith sample, ht(xi)≠yiRepresenting the training error for the ith sample, Zt is a normalization factor:
Zt＝sum(D)
the second learning is carried out, and after t rounds of learning are carried out, t weak learning algorithms { h } are obtained1,h2,....,htAnd its weight { alpha }1,α2,....,αtAnd f, respectively calculating t weak classifications for the new classification dataDevice { h1(X),h2(X),....,ht(X), the final output result of the AdaBoost algorithm is as follows:
where sign (x) is a sign function.
8. The method of claim 7, wherein four monomer methods, namely a BP neural network, a GRNN neural network, an ELM extreme learning machine and an ORPC optimized regression power curve, are combined in the AdaBoost algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110846795.8A CN113610285A (en) | 2021-07-26 | 2021-07-26 | Power prediction method for distributed wind power |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110846795.8A CN113610285A (en) | 2021-07-26 | 2021-07-26 | Power prediction method for distributed wind power |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113610285A true CN113610285A (en) | 2021-11-05 |
Family
ID=78305473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110846795.8A Pending CN113610285A (en) | 2021-07-26 | 2021-07-26 | Power prediction method for distributed wind power |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113610285A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115081742A (en) * | 2022-07-22 | 2022-09-20 | 北京东润环能科技股份有限公司 | Ultra-short-term power prediction method for distributed wind power plant and related equipment |
| CN117996757A (en) * | 2024-04-07 | 2024-05-07 | 南京中核能源工程有限公司 | Distributed wind power based power distribution network scheduling method, device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102102626A (en) * | 2011-01-30 | 2011-06-22 | 华北电力大学 | Method for forecasting short-term power in wind power station |
| CN103745274A (en) * | 2014-01-08 | 2014-04-23 | 贵州电力试验研究院 | Short-term power prediction method applied to dispersed wind power |
| CN109784563A (en) * | 2019-01-18 | 2019-05-21 | 南方电网科学研究院有限责任公司 | Ultra-short-term power prediction method based on virtual anemometer tower technology |
| CN109800923A (en) * | 2019-02-22 | 2019-05-24 | 南方电网科学研究院有限责任公司 | distributed wind power generation short-term power combination prediction method |
| CN111489038A (en) * | 2020-04-14 | 2020-08-04 | 宁波金风绿能能源有限公司 | Method for improving accuracy of ultra-short-term power prediction of wind power plant |
-
2021
- 2021-07-26 CN CN202110846795.8A patent/CN113610285A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102102626A (en) * | 2011-01-30 | 2011-06-22 | 华北电力大学 | Method for forecasting short-term power in wind power station |
| CN103745274A (en) * | 2014-01-08 | 2014-04-23 | 贵州电力试验研究院 | Short-term power prediction method applied to dispersed wind power |
| CN109784563A (en) * | 2019-01-18 | 2019-05-21 | 南方电网科学研究院有限责任公司 | Ultra-short-term power prediction method based on virtual anemometer tower technology |
| CN109800923A (en) * | 2019-02-22 | 2019-05-24 | 南方电网科学研究院有限责任公司 | distributed wind power generation short-term power combination prediction method |
| CN111489038A (en) * | 2020-04-14 | 2020-08-04 | 宁波金风绿能能源有限公司 | Method for improving accuracy of ultra-short-term power prediction of wind power plant |
Non-Patent Citations (4)
| Title |
|---|
| 吴俊利等: "基于Adaboost的BP神经网络改进算法在短期风速预测中的应用基于Adaboost的BP神经网络改进算法在短期风速预测中的应用", 电网技术, vol. 36, no. 9, 30 September 2012 (2012-09-30), pages 221 - 225 * |
| 贵州电网有限责任公司编: "高原山区风电功率预测技术", 31 March 2017, 贵州大学出版社, pages: 122 * |
| 高阳;钟宏宇;葛延峰;董存;许傲然;: "基于GRNN全信息神经网络的超短期风速预测研究", 测控技术, no. 04, 18 April 2016 (2016-04-18), pages 149 - 152 * |
| 龙源电力集团股份有限公司，中国能源报社著: "寻找真实的成长：龙源电力二十年发展之路", 31 January 2013, 中国环境出版社, pages: 122 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115081742A (en) * | 2022-07-22 | 2022-09-20 | 北京东润环能科技股份有限公司 | Ultra-short-term power prediction method for distributed wind power plant and related equipment |
| CN117996757A (en) * | 2024-04-07 | 2024-05-07 | 南京中核能源工程有限公司 | Distributed wind power based power distribution network scheduling method, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Raza et al. | An ensemble framework for day-ahead forecast of PV output power in smart grids | |
| CN103268366B (en) | A kind of combination wind power forecasting method suitable for distributing wind power plant | |
| Shi et al. | Hybrid forecasting model for very-short term wind power forecasting based on grey relational analysis and wind speed distribution features | |
| Liu et al. | Wind power plant prediction by using neural networks | |
| CN102663513B (en) | Utilize the wind power combined prediction modeling method of grey relational grade analysis | |
| CN111753893A (en) | Wind turbine generator power cluster prediction method based on clustering and deep learning | |
| CN104978608B (en) | A kind of wind electric powder prediction device and prediction technique | |
| CN108388962B (en) | Wind power prediction system and method | |
| CN113657662B (en) | Downscaling wind power prediction method based on data fusion | |
| CN109274121B (en) | Wind power plant control parameter optimization method and system | |
| CN102496927A (en) | Wind power station power projection method based on error statistics modification | |
| CN111488896B (en) | Distribution line time-varying fault probability calculation method based on multi-source data mining | |
| CN103473621A (en) | Wind power station short-term power prediction method | |
| CN110766200A (en) | Method for predicting generating power of wind turbine generator based on K-means mean clustering | |
| CN113205210B (en) | Wind power plant wind speed and power prediction method, system and equipment for complex terrain and storage medium | |
| CN105389634A (en) | Combined short-term wind power prediction system and method | |
| CN103489046A (en) | Method for predicting wind power plant short-term power | |
| CN105512766A (en) | Wind power plant power predication method | |
| CN111626473A (en) | Two-stage photovoltaic power prediction method considering error correction | |
| CN105825002B (en) | A kind of wind power plant dynamic equivalent modeling method based on dynamic Gray Association Analysis | |
| CN113627674A (en) | Distributed photovoltaic power station output prediction method and device and storage medium | |
| Chen et al. | Research on wind power prediction method based on convolutional neural network and genetic algorithm | |
| CN113610285A (en) | Power prediction method for distributed wind power | |
| CN110738253A (en) | short-term wind power prediction method based on FCM and AFSA-Elman | |
| CN110991701A (en) | Wind power plant fan wind speed prediction method and system based on data fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |